sunho
88f6a41208
shader: fix type in vtst
...
It was using wrong type before.
2020-09-04 15:26:08 +09:00
sunho
7565425a30
shader: int type conversion in vpck
...
We can now remove this conversion table and use utiliy function. Once it's in right type, store function should work correctly.
2020-09-04 15:26:08 +09:00
sunho
afe8028b57
shader: int type conversion in vbw
...
The destination is uint32
2020-09-04 15:26:08 +09:00
sunho
6cb683c3f1
shader: int type conversion in sop
2020-09-04 15:26:08 +09:00
sunho
a445f17f6f
shader: int type conversion in texture fetch & fragment output
2020-09-04 15:26:08 +09:00
sunho
c4eb07d4f4
shader: add type conversion utility functions
2020-09-04 15:26:08 +09:00
sunho
432e574f90
shader: store and load int values with ivec and uvec in glsl
...
We were loading and storing using normalized float before, but this can be quite troublesome when we start to seriously implement int instructions.
1. The rounding error can result in the change in value. The error caused by rounding error assuming 0.5 ulp is +-(1/2)*2^(1-23) * 2^16 * x = +-x*1.1920929e-7*2^16 = +-0.0078125*x where x is the value we want to store. It can be serious enough to change the value after one store & load.
2. We will need a paramter that decides the final type of load that can be normalized float, float, and integer. But, we're already specifying type by op.type! Load function is given another responsibility: type conversion. This becomes more confusing in store function. The function doesn't know whether source is normalized or not, so we need another parameter. And, this parameter should mean nothing in immediate register bank, etc. aaaaa
For this reason, I'm moving to integer loading and storing using ivec & uvec. This way load and store function has less side effect. The ouput and input type of oprand is just decided by the op.type and dest_type. Instructions using load and store is in charge of type conversion now and it wouldn't be insane because they have more context.
2020-09-04 15:26:08 +09:00
sunho
d81fd7a62b
shader: fix ubo not binding
2020-08-29 09:20:00 +09:00
sunho
a8bd23bbcc
shader: uniform buffers inside program_input
2020-08-29 06:10:15 +09:00
sunho
cee3ecbc50
shader: add comments
2020-08-29 06:10:15 +09:00
sunho
8b176ce65b
shader: fix non ubo uniforms
...
We copied it to pa before instead of sa.
2020-08-29 06:10:15 +09:00
sunho
d4f2b676f1
shader: refactor gxp parameter parsing
...
We need parsed parameters in many applications such as analyzer/fuzzer/debugger. I separated parsing part from recompiler. In the way, I fixed some bugs that are not apparent because of parsing code mixed with spirv generation code.
2020-08-29 06:10:15 +09:00
sunho
caf102d956
shader: add missing F32 in get_parameter_type_store_and_name
2020-08-29 06:10:15 +09:00
sunho
3c3ae5d94f
shader: add translate_generic_type
2020-08-29 06:10:15 +09:00
sunho
88500bad81
shader: refactor uniform buffer parsing
2020-08-29 06:10:15 +09:00
sunho
f05cd22127
shader: refactor get_parameter_type_and_name
2020-08-29 06:10:15 +09:00
sunho
1bbc744e4a
shader: use swizzle decoing function instead of constant
2020-08-29 06:10:15 +09:00
sunho
4b87fb6b3e
shader: implement dual frcp
2020-08-29 06:05:30 +09:00
sunho
db81fa0450
shader: output of dot product is scalar
2020-08-29 06:05:30 +09:00
sunho
096f6b6342
shader: implement dual frsq
2020-08-29 06:05:30 +09:00
sunho
f332307b0e
shader: initialize vertex outputs to 0
2020-08-29 06:05:23 +09:00
sunho
83aeb849fa
shader: use constants for register array size
2020-08-29 06:05:23 +09:00
sunho
f9ae3cb12a
shader: vldst should not match special
2020-08-20 05:52:20 +09:00
sunho
32a734d741
shader: move int and illegal instructions to separate files
2020-08-20 05:52:20 +09:00
sunho
9b93ea736f
shader: log unimplemented and illegal opcodes
2020-08-20 05:52:20 +09:00
sunho
d3c4ec4ed0
shader: add dummy decoders for unimplemented instructios & order instructions according to opcode1
2020-08-20 05:52:20 +09:00
sunho
712275e0b3
shader: rename some instructions
2020-08-20 05:52:20 +09:00
KorewaWatchful
83e7cb5509
shader: account for parameter-less buffers
...
Co-authored-by: 1whatleytay <1whatleytay@hdsb.ca>
2020-08-20 04:00:36 +09:00
sunho
c6cdb9fef4
shader: implement imad
2020-08-18 06:14:23 +09:00
sunho
468465470f
shader: consider Immediate in operand_to_str
2020-08-18 06:14:23 +09:00
sunho
f4bb756b3e
shader: implement integer load and store
2020-08-18 06:14:23 +09:00
sunho
04791a3170
shader: add imad decoder
2020-08-18 06:14:23 +09:00
sunho
e9ebe16084
shader: Implement unpacking & packing one of int and uint values
2020-08-17 18:59:28 +09:00
sunho
167c21363e
shader: swizzle cycling in uint8 vpck
2020-08-15 22:51:28 +09:00
sunho
5de2aa1751
shader: do not bitcast integer types
2020-08-15 22:51:28 +09:00
sunho
31112ac530
shader: use normalized value as default for unpacked uint8 vector
2020-08-15 22:51:28 +09:00
sunho
096e0d92cb
shader: update TODO
2020-08-15 22:51:28 +09:00
sunho
cea6d5b703
shader: consider normalization of uint8 input, output and texture
2020-08-15 22:51:28 +09:00
sunho
576f04085e
shader: cast uint8 bytes in differnet way in vpck
2020-08-15 22:51:28 +09:00
sunho
67cf704f7c
shader: use uint8 in sop2
2020-08-15 22:51:28 +09:00
sunho
fabee5907b
shader: implement pakcing & unpacking uint8
2020-08-15 22:51:28 +09:00
sunho
29797bca47
shader: add unwrap_type utility function
2020-08-15 22:51:28 +09:00
sunho
7ac71a203f
shader: log scale in vpck
2020-08-15 22:51:28 +09:00
sunho
a7e3897380
shader: stub uint8 unpack & pack
2020-08-15 22:51:28 +09:00
sunho
4b22c0606d
shader: log sop2
2020-08-15 22:51:28 +09:00
sunho
4283dbe409
shader: support uchar sampler
2020-08-15 22:51:28 +09:00
sunho
89e9acd728
shader: support fmin and fmax in sop2
2020-08-15 22:51:28 +09:00
sunho
f15c84ca77
shader: correct predicate field of sop2
2020-08-15 22:51:28 +09:00
sunho
447a8f8f1a
shdaer: adjust pa offset in non native color
2020-08-02 10:35:49 +09:00
Sunho Kim
3f2c8b69f3
shader: dump shader before crashing
2020-08-01 05:34:08 +09:00
